mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* Could hd-drivers and buffer algorithm be hardware?
@ 2001-04-17 15:20 John Nilsson
  2001-04-17 16:56 ` Tomas Telensky
  0 siblings, 1 reply; 2+ messages in thread
From: John Nilsson @ 2001-04-17 15:20 UTC (permalink / raw)
  To: linux-kernel

The idea is as follows.

Design a hardisk controller that would take care of all harddrive and block 
device managment and provide a virtual storage area to the OS. This way all 
the kernel would have to worry about is a virtual harddrive and how to fech 
and write data from and to it. Buffering, and read/write optimization would 
be taken care of by the controller.

The controller would have a proccessing unit, its own memory, and a chip to 
compress/decompress data.
The compression chip would filer all read and written data so that the 
actual amount of data that is read and written to disk is compressed, this 
way increasing disk space, and speed up disk read/writes.
The memory is a SDRAM DIMM that could be upgraded for more memmory, needed 
if you would want to add more physical disks or just make room for mor disk 
cache/ buffers.
The chip would take care of diskdriver issues, raid, buffering, and 
diskplacement optimization. For instance it could make a note of what files 
is usually read together and frequently, placing them close to eachother and 
on the outer tracks of the hardrives if they are big, or more generally in 
the middle of the used drivespace to optimize head movements...

>From the kernel side you would have a singel gigantic ultra fast hardrive, 
and the disk drivers would be loaded inte the chip bios on installation 
time. Further the buffering algorithms would also be loade inito the chip 
bios on installation time to decrease the mainCPU time of kernel code.


I'm just a curious computer nerd, but tell me is it a good idéa?

/John Nilsson
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Could hd-drivers and buffer algorithm be hardware?
  2001-04-17 15:20 Could hd-drivers and buffer algorithm be hardware? John Nilsson
@ 2001-04-17 16:56 ` Tomas Telensky
  0 siblings, 0 replies; 2+ messages in thread
From: Tomas Telensky @ 2001-04-17 16:56 UTC (permalink / raw)
  To: John Nilsson; +Cc: linux-kernel

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: TEXT/PLAIN; charset=X-UNKNOWN, Size: 2359 bytes --]



On Tue, 17 Apr 2001, John Nilsson wrote:

> The idea is as follows.
> 
> Design a hardisk controller that would take care of all harddrive and block 
> device managment and provide a virtual storage area to the OS. This way all 
> the kernel would have to worry about is a virtual harddrive and how to fech 
> and write data from and to it. Buffering, and read/write optimization would 
> be taken care of by the controller.
> 
> The controller would have a proccessing unit, its own memory, and a chip to 
> compress/decompress data.
> The compression chip would filer all read and written data so that the 
> actual amount of data that is read and written to disk is compressed, this 
> way increasing disk space, and speed up disk read/writes.
> The memory is a SDRAM DIMM that could be upgraded for more memmory, needed 
> if you would want to add more physical disks or just make room for mor disk 
> cache/ buffers.
> The chip would take care of diskdriver issues, raid, buffering, and 
> diskplacement optimization. For instance it could make a note of what files 
> is usually read together and frequently, placing them close to eachother and 
> on the outer tracks of the hardrives if they are big, or more generally in 
> the middle of the used drivespace to optimize head movements...
> 
> >From the kernel side you would have a singel gigantic ultra fast hardrive, 
> and the disk drivers would be loaded inte the chip bios on installation 
> time. Further the buffering algorithms would also be loade inito the chip 
> bios on installation time to decrease the mainCPU time of kernel code.
> 
> 
> I'm just a curious computer nerd, but tell me is it a good idéa?

I don't think. The expensive RAM would not be efficiently used. OTOH, the
cheap HDD need not to be compressed, how would you recover the data? You
also mix in the filesystem layer (relocating files), it would be hardly
possible.

  Tomas Telensky

> 
> /John Nilsson
> _________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2001-04-17 16:56 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-04-17 15:20 Could hd-drivers and buffer algorithm be hardware? John Nilsson
2001-04-17 16:56 ` Tomas Telensky

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®